While the incidence of most major cancers appear stable or declining, the incidence of esophageal adenocarcinoma (EA) has increased 6-fold in the U.S. since the 1970s for reasons unknown. Known risk factors for EA are cigarette smoking, obesity, and reflux esophagitis; however, these factors do not completely explain this emerging public health burden. It is urgent to improve our understanding of EA etiology and find out ways to prevent this disease. We hypothesize that the oral microbiome contributes to the development of esophageal adenocarcinoma based on several lines of evidence from our preliminary case control studies: i) there is a global alteration of foregut microbiome in esophageal adenocarcinoma with the strongest changes found in the oral microbiome, ii) commensal oral bacteria are capable of activating or degrading carcinogens in cigarette smoke, and iii) the oral microbiome has potential predictive power for development of oral cancer in the NIH-PLCO cohort. These data suggest that alterations of oral microbiome could contribute to cancer development locally (oral cavity) and downstream (esophagus). The PLCO pre-diagnostic buccal cell samples are an indispensable resource in the next step of esophageal cancer microbiome study because: These PLCO pre-diagnostic buccal cell samples are an indispensable resource in the next step of esophageal cancer microbiome study because: i) We have evaluated the quality of the PLCO samples and found that the samples collected years ago still contain sufficient microbial genetic materials needed for oral microbiome profiling (see section C2), ii) We have compared microbiome alterations in esophageal diseases in the mouth, esophagus, and stomach and found the strongest phenotype association was in the mouth (see section C1), iii) use of the samples collected prior to cancer diagnosis allows us to examine whether microbiome changes precede to cancer development, and iv) use of already collected samples and diet and demographic information in this high quality cohort makes this study cost efficient. We propose a prospective study to resolve the temporal order and etiological relationship between changes in microbiome and development of EA using pre-diagnostic buccal cell samples collected in the NCI-PLCO trial and American Cancer Society Cancer Prevention Study II (ACS-CPS II). This will be a nested case control study including 106 incident esophageal adenocarcinoma cases and 212 controls. We will identify bacterial taxa (Aim 1), bacterial genes/pathways (Aim 2), and non- bacterial microorganisms (viruses, fungi, Archaea) (Aim 3) associated with risk for development of esophageal adenocarcinoma. Using this information, we will construct a panel of integrated biomarkers to predict risk of development of esophageal adenocarcinoma. Knowledge gained from this study could add a new dimension, i.e., microbial risk factors, to our understanding of the etiology and recent surge of esophageal adenocarcinoma. If identified, these pre-diagnostic microbial biomarkers could be used to stratify at-risk subjects into high and low risk groups for more efficient screening and early detection of EA. Normalization of the microbiome in at-risk patients could become a new means for preventing esophageal adenocarcinoma.